What does “representation” mean in a society with rapidly changing demographics?
One of the most significant changes that has occurred in American social justice debates over the past few decades has been a shift in the justifications that are offered for race-based affirmative action programs. Rather than attempting to eliminate discrimination from the procedures that are followed in various institutions, the focus has shifted to producing outcomes that “represent” the population. This is summed up best by the expressed ambition to create institutions that “look like America.”
(Perhaps the most dramatic example of this can be seen in the campaign to end blind auditions in U.S. symphony orchestras, in order to increase the number of black musicians. Since no one could seriously maintain that black Americans need special help breaking into the music industry, and since the blind audition eliminates any procedural concern over discrimination, the only possible justification for such an initiative is the purely outcome-based one – what proponents call “representation” – that every institution in society should mirror the demographics of the population as a whole.)
Philosophers sometimes refer to this way of thinking about equality as telic egalitarianism (which is the view that equality in some dimension is a good-making property of outcomes). The shift toward this way of thinking in U.S. racial politics received its greatest impetus from the Regents of the University of California v. Bakke Supreme Court Decision (1978), which ruled that affirmative action programs in educational admissions were of dubious constitutionality as procedural constraints, but could be defended on the grounds that promoting diversity in the classroom was a valid institutional objective, because it enhanced the quality of the educational experience. This argument encouraged Americans to think about diversity in consequentialist terms, not in terms of non-discrimination, but rather as an outcome to be achieved.
These ideas have many admirers and imitators in Canada. And yet they are much more difficult to apply, because the question of what representation means in this country is more complex than it is in the U.S. The first challenge arises from the simple fact that most Canadians have no idea what the demographics of the country are, and because they consume so much American media, often have wildly erroneous beliefs. The problem is worse with people who live in cities, who often fail to appreciate how different the urban population is from the rural. The second challenge arises from the fact that the demographic profile of the population is changing so radically – after all, what does it mean to create institutions that “look like Canada,” when the population is changing at a rate of over 1% per year? The final, most subtle challenge arises from the fact that the demographic profile of the population has been changing for a while, in non-random ways, and so different age cohorts have very different demographics.
The first challenge seems a bit ridiculous but is surprisingly common. When I was younger, the biggest problem I encountered in this area was with Canadians who sought to help “under-represented” groups without actually checking to see what the demographics of the country were. I can recall one episode, back in the 1990s, when graduate students in our department came to us with a demand to create an affirmative action program for visible minorities in PhD admissions. A simple head-count revealed that every major visible minority group in Canada (with the exception of Indigenous peoples, who raise a somewhat different set of issues) was already overrepresented in our program. The students were pushing on an open door.
The problem was that they were assuming, without thinking, that the numbers in Canada must be roughly similar to those in the U.S. (e.g. where blacks make up about 13% of the population). In fact, at the time all visible minorities combined in Canada made up less than 10% of the population (with blacks at around 2%). Our department at the time did “look like Canada,” the students just hadn’t seen enough of Canada to know what Canada looks like.
That of course has changed radically in the past twenty years, as the following table from Statistics Canada indicates (drawn from this useful page):
I often wish that the people I work with would take a look at these statistics before they talk about representation. I have on more than one occasion intervened in meetings to prevent colleagues from using “Hispanic” as an affirmative action category. (This particular problem is exacerbated by the over-representation of Americans in my workplace.) Meanwhile, I have met very few who realize what a gigantic Filipino population this country has, and none who care even a tiny bit how well represented they are in Canadian higher education (much less Canadian institutions more generally). The same goes for Muslims, who are admittedly not a racial group, but are arguably one of the classes of immigrants most likely to experience discrimination (and now make up almost 5% of the Canadian population).
This brings us to the second set of problems, which has to do with the fact that Canadian demographics are in a state of rapid flux. In the year that I was born, black Canadians made up just 0.1% of the population. The black population subsequently increased to about 2% by the end of the century, and has since doubled to 4%. Approximately one-third of the current black population in Canada was born in Africa. (In the U.S., by contrast, the black share of the population has remained roughly constant.)
This sort of rapid demographic change means that how representative an institution is, with respect to the population, will be strongly influenced by such factors as the rate of turn-over in personnel. It means, for example, that there is going to be a lot of diversity in bank tellers, but a lot less diversity in bank managers; a lot of diversity among journalists, a lot less among editors, etc. Any institution that relies on internal promotion to fill its senior positions is going have highly unrepresentative leadership, simply because of this time lag. I was hired at University of Toronto during the 1990s, and most of the people that I work with, at my level of seniority, were also hired in the ‘90s. Unsurprisingly, the demographics of my cohort more closely resemble the demographics of Canada in the ‘90s than they do the present age. So there is a huge mismatch between our student population, which is minority-majority, and the faculty, which isn’t. Some of this may be due to past discrimination, but it is impossible to draw this conclusion just by pointing to the mismatch.
But wait, a critic might say, it’s not as though all of these institutions have to wait for second-generation immigrants to work their way up through the ranks. They could also be hiring older immigrants directly into these more senior positions, in a way that would preserve something closer to representation. The problem with this argument is that the Canadian immigration system does not pick people randomly to let in. It selects people with specific skills, in the hope that they will enter specific professions. As a result, the demographics in some professions (e.g. nursing) are shifting far more rapidly than in others. There is not a large pool of unemployed immigrants with advanced degrees in the humanities for us to draw upon, because (surprise, surprise) you don’t get a lot of points for having an advanced degree in the humanities.
As I have argued elsewhere, I think the whole shift to thinking about equality as an outcome to be achieved, rather than a procedural constraint on the way that individuals are treated by social institutions, is a mistake. Even though the distinction is difficult to draw in many cases, the way that American progressives have been thinking about representation gives substance to the conservative critique that egalitarians are committed to a form of social engineering that crowds out individual preference. Canadians would be foolish to follow them in this regard, but even more so, because in trying to engineer particular outcomes, they are aiming at a rapidly moving target.
The lesson to be drawn, in my view, is that we should redouble our focus on the traditional objective of eliminating discrimination, with particular focus on the procedures that are being followed to select individuals for entry and advancement in institutions, and stop worrying about whether they wind up reproducing internally all of the diversity that exists in the broader society. And to the extent that we do look at issues of representation, we should avoid doing snap-shot comparisons, but focus instead on how institutions adapt over the course of decades.
NB: Nothing that I have said here applies to representation of women, who remain roughly constant as a fraction of the population.